Download Crystal Structure of a 70 Subunit Fragment from

Cell, Vol. 87, 127–136, October 4, 1996, Copyright 1996 by Cell Press
Crystal Structure of a s70 Subunit Fragment
from E. coli RNA Polymerase
Arun Malhotra, Elena Severinova,*
and Seth A. Darst
The Rockefeller University
1230 York Avenue
New York, New York 10021
Summary
The 2.6 Å crystal structure of a fragment of the s70
promoter specificity subunit of E. coli RNA polymerase
is described. Residues involved in core RNA polymerase binding lie on one face of the structure. On the
opposite face, aligned along one helix, are exposed
residues that interact with the 210 consensus promoter element (the Pribnow box), including four aromatic residues involved in promoter melting. The
structure suggests one way in which DNA interactions
may be inhibited in the absence of RNA polymerase
and provides a framework for the interpretation of a
large number of genetic and biochemical analyses.
Introduction
Transcription is the major control point of gene expression and DNA-dependent RNA polymerase (RNAP) is
the central enzyme of transcription. The core RNAPs
from bacterial and eukaryotic cells, which are catalytically active in RNA chain elongation, are homologous
in structure and function (Allison et al., 1985; Biggs et
al., 1985; Ahearn et al., 1987; Sweetser et al., 1987;
Darst et al., 1989; Darst et al., 1991; Schultz et al., 1993;
Polyakov et al., 1995). Specific initiation of transcription
requires additional protein factors. Promoter-specific
initiation of mRNA synthesis in eukaryotes requires a
collection of basal initiation factors (TFIIB, TBP, TFIIE,
TFIIF, and TFIIH, reviewed in Roeder, 1991; Conaway
and Conaway, 1993) comprising more than a dozen
polypeptides with a total mass z750 kDa. In bacteria,
specific initiation by RNAP requires a single polypeptide
known as a s factor, which binds to core RNAP to form
the holoenzyme (Burgess et al., 1969; Travers and Burgess, 1969). One primary s factor directs the bulk of
transcription during exponential growth. Specialized, alternative s factors direct transcription of specific regulons during unusual physiological or developmental conditions (reviewed in Helmann and Chamberlin, 1988;
Gross et al., 1992). The primary and most of the alternative s factors comprise a highly homologous family of
proteins (Stragier et al., 1985; Gribskov and Burgess,
1986) with four regions of highly conserved amino acid
sequence (Figure 1; reviewed in Lonetto et al., 1992).
The primary s factor in Escherichia coli, s70 , directs
transcription from promoters characterized by two elements of consensus DNA sequence: TATAAT (the Pribnow box), centered at about 210 with respect to the
*On leave from the Institute of Molecular Genetics, Russian Academy of Sciences, Moscow, Russia.
transcription start site (11), and TTGACA, centered at
about 235. The 210 and 235 elements are usually separated by 17 base pairs of nonconserved sequence (Hawley and McClure, 1983; Harley and Reynolds, 1987).
Alternative s factors of the s70 family also direct transcription from promoters organized into 210 and 235
elements but with different consensus sequences (reviewed in Helmann and Chamberlin, 1988), leading to
the proposal that s factors themselves directly contact
the DNA at the consensus promoter elements to confer
sequence-specific recognition (Losick and Pero, 1981).
Studies demonstrating the formation of chemical crosslinks between promoter DNA and s70 support this proposal (Simpson, 1979; Park et al., 1980; Hilton and
Whiteley, 1985; Buckle et al., 1991). Stronger support
comes from genetic studies demonstrating allele-specific suppression of promoter mutations by specific mutations in the corresponding s factor (Gardella et al.,
1989; Siegele et al., 1989; Zuber et al., 1989; Daniels et
al., 1990; Waldburger et al., 1990), identifying residues
in s conserved regions 2.4 and 4.2 as specifying recognition of the 210 and 235 promoter elements, respectively. However, s70 does not bind promoter DNA in the
absence of core RNAP. Recently, specific interactions
between N-terminally truncated derivatives of s70 and
promoter DNA were indicated by competetive filter retention assays, leading to the hypothesis that the latent
DNA binding activity of s70 is inhibited by N-terminal
regions and that this inhibition is relieved upon the binding of s70 to core RNAP (Dombroski et al., 1992).
Since only the holoenzyme form of RNAP forms
transcription-competent open complexes on doublestranded DNA at promoters, and since the melted DNA
includes the 210 consensus element (Siebenlist et al.,
1980; Kirkegaard et al., 1983), it has also been suggested
that s factors are directly involved in DNA melting (Hinkle
and Chamberlin, 1972), perhaps by sequence-specific
binding to one of the DNA strands (Helmann and Chamberlin, 1988). Chemical probing of open complexes indicates strong protection of the nontemplate strand by
RNAP holoenzyme, particularly in the region around the
210 consensus element (Siebenlist et al., 1980; Buckle
and Buc, 1989; Chan et al., 1990). Moreover, cross-links
between s and promoter DNA in open complexes form
preferentially with the nontemplate strand (Simpson,
1979; Park et al., 1980; Hilton and Whiteley, 1985; Buckle
et al., 1991). Interaction of s with the nontemplate strand
would maintain the melted region of DNA in the open
complex while allowing access of the RNAP catalytic
machinery and nucleotide substrates to the template
strand.
Despite the central importance of s factors in the
control of bacterial gene expression, fundamental questions regarding the mechanism of action, the regulation,
and the role of s in such processes as promoter recognition, promoter melting, and promoter clearance, remain
unanswered. This is due, in large part, to a total lack of
structural information. We now report the X-ray crystal
structure of a s70 fragment containing conserved region
2, which is the most highly conserved region in the s 70
Cell
128
Figure 1. Conserved Regions of the s70 Family of s Factors
The bar at top represents the E. coli s70 primary sequence with amino acid numbering shown above. Evolutionarily conserved regions are
labeled below the bar and colored either gray or as follows: 1.2, red; 2.1, green; 2.2, yellow; 2.3, cyan; 2.4, orange. Expanded underneath is
a sequence alignment showing regions 1.2 to 2.4 of 11 primary s factors taken from the more extensive alignment of Lonetto et al., 1992, but
also including the primary s from E. coli C (Christie and Cale, 1995). The sequences are presented in one-letter amino acid code and are
identified by a three-letter species code (ANA, Anabaena; BSU, Bacillus subtilis; CTR, Chlamydia trachomatis; ECO, E. coli; MXA, Myxococcus
xanthus; PAE, Pseudomonas aeruginosa; SAU, Staphylococcus aureaus; SCE, Streptomyces coelicolor; STY, Salmonella typhimurium) and
a four-letter s abbreviation. Numbers at the beginning of each line indicate amino acid positions relative to the start of each mature protein
sequence. Numbers at the top indicate the amino acid position in E. coli K-12 s 70. Amino acid similarity >50% in the full alignment of Lonetto
et al. (1992) is indicated by a gray background, gaps are indicated by dotted lines. Groups of residues considered similar are: ST, RK, DE,
NQ, FYW, and ILVM. Helices are indicated above the sequences as rectangles; loops, as a solid black line. The dashed parts indicate disordered
regions that were not modeled. The patterned helices form intramolecular, antiparallel coiled-coil dimers with the like-patterned helix. Indicated
below the sequences are the locations of deletions (lines) or point mutations (dots) that affect core RNAP binding (green; Lesley and Burgess,
1989; Shuler et al., 1995), promoter melting (blue; Jones and Moran, 1992; Juang and Helmann, 1994; deHaseth and Helmann, 1995; Juang
and Helmann, 1995), or 210 consensus element recognition (orange; Kenney et al., 1989; Siegele et al., 1989; Zuber et al., 1989; Daniels et
al., 1990; Waldburger et al., 1990; Tatti et al., 1991). Also denoted is a short region of homology between s 70 and s54, an alternative s factor
that is not a member of the s 70 family, in which point mutations occur that cause loss of s 54 binding to core RNAP (green bracket; Tintut and
Gralla, 1995).
family (Lonetto et al., 1992). Conserved region 2 has
been implicated in core RNAP binding (Lesley and Burgess, 1989), recognition and binding of the 210 promoter element (Siegele et al., 1989; Zuber et al., 1989;
Daniels et al., 1990; Waldburger et al., 1990), and promoter melting (Helmann and Chamberlin, 1988; Juang
and Helmann, 1994; Rong and Helmann, 1994). The
structure reveals a new fold with a topology unlike other
known double- or single-stranded DNA binding proteins.
The structure also provides insight into the role of s
in promoter melting, indicates one way in which DNA
interactions could be inhibited in free s, and provides
a framework for the interpretation of a large number of
genetic and biochemical analyses as well as for the
design of future studies.
Results and Discussion
Crystallographic Analysis
Limited proteolysis, N-terminal sequencing and mass
spectrometry were used to determine the domain organization of E. coli s70 (E. S. et al., submitted). One of the
domains (s70 residues 114 to 448, hereafter referred to
as s702) contains a part of conserved region 1.2 and all
Structure of a s 70 Subunit Fragment
129
Figure 3. Core RNAP Binding Surface of s702
Figure 2. Structure of s702
(a) Schematic diagram of the secondary structure of s 702. Helices
are shown as cylinders. Three disordered regions that were not
modeled are indicated (not to scale) as dotted lines. These are
residues 168–172, 192–211, and 238–241. Conserved regions 1.2,
2.1, 2.2, 2.3, and 2.4 of the s70 family are color coded as in Figure 1. The nonconserved region inserted between conserved regions 1.2 and 2.1 is colored gray.
(b) RIBBONS (Carson, 1991) diagram of the three-dimensional structure of s702. Color coding is as before. Unmodeled, disordered regions are indicated (not to scale) as dotted lines: (left) similar view as
Figure 2a, perpendicular to the horizontal, pseudo-2-fold symmetry
axis; (right) view along the pseudo-2-fold symmetry axis.
but the C-terminal 8 residues of conserved region 2
(Figure 1). The 39 kDa domain binds core RNAP competetively with s70 , and the complex thus formed specifically binds single-stranded DNA containing the 210
consensus sequence of the nontemplate strand. The
structure was solved using multiwavelength anomalous
diffraction (MAD) data collected from a single crystal
of selenomethionyl-substituted protein (Hendrickson,
1991). The current model, refined to 2.6 Å resolution,
contains residues 114–446 of s 70, along with one
N-terminal methionine residue from the expression vector. Three N-terminal residues from the vector, the s70
C-terminal residues 447–448, and a total of 29 internal
residues in 3 loops are disordered and could not be
modeled (Figure 1).
Overall Topology and Organization
The structure of s702 (Figure 2), which consists entirely
of helices and connecting loops, can be divided conceptually into three substructures comprising two structural
motifs. One motif consists of an antiparallel a-helical
The backbone trace is displayed as a white tube, showing the cluster
of four helices comprising the conserved regions. A partially transparent, solvent-accessible surface encloses the structure. The large
black labels denote the helices. The view is directly at the face
containing helices 12a and 12b, comprising conserved region 2.1.
The kink between these helices is centered about Asn-383. Helices
1 and 12a form an intramolecular, antiparallel coiled-coil dimer, only
part of which is shown. Some selected, conserved hydrophobic
residues are shown in yellow. Hydrophobic a and d positions of the
coiled-coil heptad repeat are occupied by Ile-119, Ile-123, Ala-375,
and Met-379. The green backbone and side chains (residues 380–
385) denote a region known from mutagenesis studies to be important for core RNAP binding (see Figure 1). Shown also are other
highly conserved residues that we speculate may be involved in
core RNAP binding. These are, first, an adjacent patch of conserved,
solvent-exposed hydrophobic residues (Leu-384, Ile-388, Phe-401,
Leu-402, Ile-405, shown in yellow) that are suggestive of a protein
interaction surface and, second, three other highly conserved, exposed residues (shown in blue) (generated using the program
GRASP; Nicholls et al., 1991).
coiled-coil dimer (Figure 3) with a helical bundle at one
end. This motif is roughly repeated twice, giving rise to
a “V”-shaped structure with pseudo-2-fold symmetry
(Figures 2A and 2B) that is not detectably reflected in the
sequence. The second motif comprises a small helical
domain that is situated at the vertex of the V. The width
of the V is z50 Å, and the distance separating the ends
of the arms is z75 Å. In the perpendicular direction, the
molecule is z30 Å thick.
The conserved regions (colored) are clustered together and closely associated in the tertiary structure
(Figure 2). The rest of the structure (gray) comprises a
large insertion between conserved regions 1.2 and 2.1
(up to 250 amino acids in the primary s of Pseudomonas
aeruginosa, 245 in E. coli s70 ), present only in some
primary s factors but nonconserved in both sequence
and length (Figure 1). The C-terminus of conserved region 1.2 is close enough to the N-terminus of conserved
region 2.1 to be linked by only a few residues, suggesting
how the conserved regions could constitute a stably
Cell
130
folded domain with essentially the same structure in the
absence of the nonconserved region.
The clustering and tertiary fold of the conserved regions appears to be determined primarily by a tightly
packed hydrophobic core (Figures 4A and 5). As might
be expected, residues contributing to this hydrophobic
core are some of the most highly conserved of all s
factors, including several absolutely conserved residues
(Val-387, Gly-408, Gly-411, Leu-412, Ala-431, and Ile435). The absolutely conserved glycine residues occur
at the points of closest approach between helices 12a
and 13 (Gly-408) and helices 13 and 14 (Gly-411). Overall,
at the 17 positions contributing to the hydrophobic core,
the residues found at each position in E. coli s70 are
93% identical when compared with primary s factors
(77% in all s factors), and hydrophobic residues are
100% conserved in primary s factors (97% in all s factors). Helix 13 (essentially region 2.2, yellow in Figure
2), being sandwiched between the helices constituting
the other conserved regions, contributes the most residues to the hydrophobic core. This helps explain why
region 2.2 is the most highly conserved region in s factors (Lonetto et al., 1992).
Core RNA Polymerase Binding
A number of observations indicate that conserved region 2.1 is critical for high affinity binding of s to core
RNAP (Figure 1). Deletion analysis has identified a region
of s70 , residues 361–390, including most of region 2.1,
that seems to be necessary and sufficient for core RNAP
binding (Lesley and Burgess, 1989). This is consistent
with a similar analysis investigating core RNAP binding
of E. coli s 32 truncation mutants (Lesley et al., 1991).
Mutations in region 2.1 of B. subtilis s E cause defects
in binding to both B. subtilis and E. coli core RNAP
(Figure 1) (Shuler et al., 1995). PCR mutagenesis was
used to identify specific amino acids important for core
RNAP binding within E. coli s 54, an alternative s factor
that is not a member of the s70 family (Wong et al., 1994;
Tintut and Gralla, 1995). The largest concentration of
mutations causing loss of core RNAP binding fell within
or just adjacent to a short stretch of residues that bears
resemblance to s 70 residues 381–385 within region 2.1.
Moreover, the sequence through this region of s54 is
arranged in a heptad repeat characteristic of coiledcoils, as is found in this region of the s70 2 structure.
Conserved region 2.1 (green in Figure 2) forms two a
helices, the C-terminal part of helix 12a and helix 12b,
with an z458 kink between them, centered about Asn383 (Figure 3). Helix 12a makes hydrophobic coiled-coil
interactions with helix 1 while both helices 12a and 12b
contribute to the hydrophobic core formed by the cluster
of conserved region helices (Figures 4A and 5). On the
opposite face, the helices are solvent exposed. Some
of the residues that appear to be important for core
RNAP binding are illustrated in Figure 3. These residues
include or fall very near the kink between helices 12a
and 12b, suggesting that this structural feature may be
important for core RNAP binding. Adjacent to this feature is a cluster of conserved, hydrophobic residues
that form a solvent-exposed hydrophobic patch that is
suggestive of a protein-binding surface and may also
be involved in s–core RNAP interactions (Figure 3). Also
denoted in Figure 3 are nearby solvent-exposed but
highly conserved residues (Glu-114, Arg-374, Lys-392).
Because these residues do not appear to make intramolecular contacts contributing to the s702 structure, their
location and conservation suggest they may also be
involved in core RNAP contacts.
Pribnow Box Recognition and Melting
A number of studies using a variety of primary and alternative s factors from E. coli and B. subtilis (summarized
in Figure 1) have identified site-specific mutations within
region 2.4 that suppress single-base mutations within
the 210 promoter element (Figure 1). All of these studies
converge on the conclusion that residues corresponding
to 437 and 440 of s 70 (unless otherwise specified, all
amino acid numbering refers to E. coli K-12 s70 ) suppress
specific base changes at the promoter position corresponding to 212 (Kenney et al., 1989; Siegele et al.,
1989; Daniels et al., 1990; Waldburger et al., 1990). Specifically in E. coli s70 , substitution of Gln-437 with His
(Gln-437-His) causes a substantial increase in activity
from mutant promoters having a T-to-C substitution at
212 (Waldburger et al., 1990), and a Thr-440-Ile substitution increases activity from mutant promoters having
T-to-C or -G substitutions at 212 (Siegele et al., 1989).
Although the 213 position is not conserved in the 210
element recognized by s 70, many alternative s factors
recognize 210 consensus elements that extend upstream to 213 or even further. Similar analyses for two
of these s factors, sH and sE of B. subtilis, have demonstrated that substitution of the residue corresponding
to 441 of s70 suppresses specific base changes at 213
(Zuber et al., 1989; Tatti et al., 1991). The spacing of
these residues, along with the presence of three absolutely conserved hydrophobic residues at 435, 439, and
443 (Figure 1) led to the proposal that this region of s
factors forms an amphipathic helix with exposed residues at positions 437 and 440 directly contacting the
212 base, while the residue corresponding to position
441 contacts the 213 base (Kenney et al., 1989; Daniels
et al., 1990; Waldburger et al., 1990). Region 2.4 does
indeed form an amphipathic a helix (helix 14), with the
conserved hydrophobic residues contributing to the hydrophobic core formed by the conserved regions, while
the residues implicated in 210 element recognition are
solvent-exposed on the opposite face of the helix (Figures 4 and 5).
A role for conserved region 2.3 in promoter melting
has been proposed (Helmann and Chamberlin, 1988)
based on its proximity to region 2.4 (the 210 recognition
region) and its high proportion of conserved aromatic
and basic residues, which might serve to stack with
exposed bases and neutralize the DNA phosphate backbone, as in single-stranded nucleic acid–binding proteins (Nagai et al., 1990; Shamoo et al., 1995). This model
has been tested with B. subtilis s A, which is 83% identical with E. coli s70 in region 2.3 (100% homologous since
the only substitutions are conservative), by substituting
each of the seven conserved aromatic residues in region
2.3 (Figure 1) with alanine (Juang and Helmann, 1994).
The results of this study are interesting to consider in
light of the s702 structure. Four of the substitutions (in
E. coli s70 numbering), Tyr-425, Tyr-430, Trp-433, and
Structure of a s 70 Subunit Fragment
131
Figure 4. DNA Interaction Surface of s 702
(A) Stereo RIBBONS (Carson, 1991) diagram of the cluster of four helices comprising the conserved regions. The view is 1808 about a vertical
axis from the view of Figure 3. Helix 14, containing part of conserved region 2.3 and conserved region 2.4, runs nearly horizontally across the
middle of the picture. Shown in yellow are residues that comprise the conserved hydrophobic core (Ile-119, Ile-123, Ala-375, Met-379, Val380, Val-387, Ala-391, Leu-399, Leu-404, Leu-412, Ala-415, Val-416, Phe-419, Phe-427, Ala-431, Ile-435, Ile-439, Ile-443). Other residues are
shown in color as follows: cyan, exposed conserved aromatic residues from region 2.3, important for promoter melting; orange, residues
known to interact with the 212 position of the 210 consensus element; blue, conserved basic residues flanking the promoter recognition
and promoter melting residues that may be involved in DNA phosphate backbone interactions.
(B) Likely orientation of helix 14/nontemplate DNA strand interactions. The backbone of helix 14 is shown as a coil with the solvent-exposed
face of the helix facing down. The a-carbon positions of residues important for promoter recognition or melting are indicated. Schematically
illustrated below is the nontemplate strand sequence of the 210 consensus element. Interactions between specific residues and bases
determined from genetic or biochemical studies are indicated by dashed lines. The interaction indicated between the residue at position 441
and the 213 position is not specific in the case of s70 (the 213 position is not conserved in the 210 element recognized by s70) but is indicated
from genetic studies on alternative s factors that recognize 210 elements with a conserved 213 position (Daniels et al., 1990).
Trp-434 (cyan in Figures 4A and 5), cause impaired DNA
melting even when the promoter is saturated with RNAP
(Juang and Helmann, 1994). Consistent with melting defects, the impairment can be overcome by using supercoiled templates or by raising the reaction temperature (Aiyar et al., 1994; Juang and Helmann, 1994).
Several of these mutations exert a trans-dominant lethal
effect in vivo (Rong and Helmann, 1994). In addition, a
substitution in an alternative s factor at the position
corresponding to 434 also results in what appears to
be a melting defect (Jones and Moran, 1992). Each of
these conserved aromatic residues is solvent exposed
and on the same face of helix 14 as the region 2.4
residues implicated in 210 element recognition (Figure
4). Two other substitutions, Phe-419 and Phe-427, exhibit properties that suggest these mutants are unable
to fold properly. Consistent with this result, these two
residues are buried and play important roles in forming
the conserved hydrophobic core (Figures 4A and 5).
Substitution of Tyr-421 has little to no effect in vivo and
in vitro, and in the s702 structure this residue does not
participate in the hydrophobic core and is not in a position to participate in DNA interactions since it faces
the opposite side of the structure as the 210 element
recognition and melting residues.
Potassium-permanganate footprinting indicates that
promoter melting nucleates within the 210 element
around the 211/210 position. Interestingly, the Tyr-430
Cell
132
Figure 5. Potential Autoinhibition of DNA Binding
RIBBONS (Carson, 1991) diagram showing a view of the conserved
region helices, z908 about a vertical axis from the view of Figure 4.
Helix 14 is viewed from the C-terminal end nearly down its axis.
Selected conserved residues are color coded as follows: yellow,
residues comprising conserved hydrophobic core; green, residues
in region important for core RNAP binding; cyan, aromatic residues
important for promoter melting; orange, residues important for recognition of the 212 position of the 210 consensus element. The
location of core binding and DNA binding determinants on opposite
sides of the structure is noted. Illustrated schematically are the
sequence (in single-letter amino acid code) and charge of the 20residue, disordered acidic loop (residues 192–211).
cleft might sterically inhibit DNA interaction and would
also repel the negatively charged DNA electrostatically.
This would help to explain why s70 and also s70 2 do
not interact with DNA in the absence of core RNAP,
consistent with studies implicating regions of s70 N-terminal of region 2 in inhibiting DNA interactions (Dombroski et al., 1992, 1993). Since many s factors lack the
acidic loop (Figure 1), it cannot be the only mechanism
by which specific interactions of s factors with promoter
DNA are inhibited in the absence of core RNAP, also
consistent with the studies of Dombroski et al., 1992.
Since the s70 2-core RNAP complex binds specifically to
a single-stranded DNA oligo with the nontemplate 210
consensus sequence, a mechanism involving substantial conformational changes of s and/or neutralization
of the acidic loop upon core RNAP binding must exist
for the autoinhibition of DNA binding from the acidic
loop to be overcome.
and Trp-433 substitutions appear to be defective in the
nucleation of melting (Juang and Helmann, 1995), suggesting that these residues interact with the bases at
the 211 and/or 210 positions. Thus, residues implicated
in interacting with the 213, 212, and 211/210 bases
are aligned along one face of helix 14. Flanking this
row of aligned residues on top and bottom are several
conserved, positively charged arginine and lysine residues (Figure 4A) that could interact with the negatively
charged phosphate backbone of the DNA. Cross-linking
and chemical probing experiments (discussed above)
suggest that s interacts primarily with the nontemplate
strand in the open promoter complex. Furthermore, a
complex between core RNAP and s70 2 specifically binds
a single-stranded DNA oligo with the 210 consensus
sequence of the nontemplate strand (E. S. et al., submitted). Thus, we propose that in the open promoter complex, the single-strand region of the nontemplate strand
interacts with the residues of region 2.3 and 2.4 in the
manner and orientation schematically illustrated in Figure 4B.
Relationship to Eukaryotic Factors
Putative sequence relationships have been suggested
between s conserved region 2 and various eukaryotic
basal transcription factors such as TATA-binding protein (TBP; Horikoshi et al., 1989), TFIIF or RAP30/74
(Sopta et al., 1989), TFIIB (Ha et al., 1991), subunits of
TFIIE (Ohkuma et al., 1991; Sumimoto et al., 1991), and
the mitochondrial RNAP factor MTF1 (Jang and Jaehning, 1991). The proposed similarity between s 70 conserved region 2 and TBP is particularly intriguing considering that each protein recognizes essentially the same
DNA sequence. However, none of these sequence alignments are statistically significant based on the criteria
of Lonetto et al., 1992. Indeed, considering s 70 region 2
and TBP in light of their structures, the relevant region
of s70 (residues 417–457, conserved regions 2.3 and 2.4)
is mostly an a helix, while the relevant region of TBP
(residues 129–170 of Arabidopsis thaliana TBP) is a short
a helix and three strands of antiparallel b sheet (Nikolov
et al., 1992). Thus, this proposed sequence relationship
does not appear to be meaningful, and functional evidence will be required to establish the significance of
the other noted similarities. In one case, the proposed
similarity between the RAP30 subunit of RAP30/74
(TFIIF) and the core binding region 2.1 of s factors (Sopta
et al., 1989) is supported by the finding that RAP30/74
binds E. coli core RNAP competetively with s 70 (McCracken and Greenblatt, 1991). It is also interesting to
note that bg, the TFIIF homolog derived from rat liver,
binds to RNAP II and inhibits its nonspecific DNA binding, which is also a function attributed to s factors (Conaway and Conaway, 1990).
Inhibition of DNA Binding
The core RNAP binding determinants and DNA binding
determinants of s 702 face opposite sides of the structure
(Figure 5). The residues implicated in 210 recognition
and melting all face into a cleft-like feature. However,
potentially occupying this cleft is a highly acidic stretch
of residues from 188 to 209. In this stretch of 22 residues,
18 are negatively charged (Figure 1). Most of this acidic
loop (192–209) is disordered. The presence of this highly
acidic loop within or near the apparent DNA-binding
Conclusion
The crystal structure of an E. coli RNAP s70 fragment
reveals an entirely helical, “V”-shaped protein with all
the regions of conserved primary sequence closely associated with one another. Residues known to interact
with core RNAP (Lesley and Burgess, 1989; Shuler et
al., 1995) face one side of the structure, while on the
opposite side, residues important for promoter recognition (Siegele et al., 1989; Waldburger et al., 1990) and
four conserved aromatic residues involved in promoter
Structure of a s 70 Subunit Fragment
133
Table 1. Summary of the Crystallographic Analysis
Diffraction Data
Native
Se-Met
Se-Met
Se-Met
Se-Met
Wavelength (Å)
Resolution (Å)
Total reflections
Unique reflections
Completeness (%) [last shell]
I/s [last shell]
Rsym (%) [last shell]
0.9663
2.6
76887
13160
94.9 [78.3]
15.1 [2.5]
6.3 [27.0]
l1 5 0.9879
2.8
58230
10267
92.3 [65.7]
17.4 [3.6]
6.8 [17.5]
l2 5 0.9793
2.8
58476
10229
92.6 [67.9]
16.6 [3.2]
7.6 [19.6]
l3 5 0.9791
2.8
58421
10221
92.7 [68.0]
16.1 [3.0]
8.8 [22.5]
l4 5 0.9686
2.8
59371
10327
93.2 [67.6]
16.0 [2.9]
8.3 [20.2]
—
—
10
1.13
10
—
10
0.55
10
1.11
Phasing statistics
Number of sites
Phasing power
Overall figure of merit: 0.591
Refinement
Resolution (Å): 6-2.6
Reflections (|F | . 2s|F |): 11376
Number of atoms in model: 2593
Number of solvent molecules: 111
R-factor (%): 21.8
Free R-factor (%): 31.5
RMS bond-length (Å): 0.012
RMS bond angle (degrees): 1.637
Rsym 5 S|I 2 ,I.|/SI, where I 5 observed intensity, ,I. 5 average intensity from multiple observations of symmetry related reflections.
Phasing power 5 root mean square (|FH|/E), where |FH| 5 heavy-atom structure factor amplitude and E 5 residual lack of closure. RMS bond
lengths and angles are the deviations from ideal values.
melting (Juang and Helmann, 1994) are aligned along
the solvent-exposed face of a single helix. The promoter
recognition and melting helix is surrounded on top and
bottom by conserved basic residues that may interact
with the DNA phosphate-backbone. Finally, we suggest
that a disordered loop of acidic residues may sterically
and electrostatically inhibit DNA interactions of s70 2 in
the absence of RNAP. Substantial conformational
changes of s 702 upon RNAP binding and/or binding and
neutralization of the acidic loop by a basic region of
RNAP would be required to overcome this inhibition.
Exposed aromatic residues surrounded by basic residues are a characteristic of single-strand nucleic acid
binding proteins, and this feature of s conserved region
2.3, noted earlier from the primary sequence (Helmann
and Chamberlin, 1988), is borne out in the structure.
Moreover, s702 promotes the specific binding of a singlestranded oligo containing the nontemplate strand sequence of the 210 consensus element by RNAP (E. S.
et al., submitted). These findings suggest that, in the
open complex between RNAP holoenzyme and promoter DNA, s functions in part as a sequence-specific
single-strand DNA binding protein. The 210 promoter
consensus element, with its T-A base steps, has a distorted structure in solution (Spassky et al., 1988). Further
distortion and unwinding would occur upon RNAP binding as a result of DNA bending (Amouyal and Buc, 1987;
Travers, 1990). The promoter melting and recognition
region of s conserved regions 2.3 and 2.4 may thus be
poised to recognize and promote melting of the highly
distorted 210 consensus element. Sequence-specific
binding of the nontemplate strand would stabilize the
transcription bubble in the open promoter complex and
leave the template strand available for the RNAP catalytic machinery.
Experimental Procedures
The identification, characterization, purification, and crystallization
of s702 (s70 residues 114–448 plus four N-terminal residues, Gly-SerHis-Met, left from the expression vector after thrombin treatment)
was as described (E. S. et al., submitted). The crystals belong to
the space group P41 21 2 with unit cell parameters a 5 b 5 79.229 Å,
c 5 134.06 Å and one molecule in the asymmetric unit. Data were
collected from frozen crystals held at 21708C, which diffracted
X-rays (produced from a rotating anode generator) strongly to
z3.5 Å and weakly to 2.9 Å resolution. Synchrotron radiation was
used to collect diffraction data to 2.6 Å resolution. We were unable
to obtain isomorphous derivatized crystals, prompting us to prepare
the protein with its 10 methionine residues (9 in the s domain and
1 from the expression vector) substituted with selenomethionine
(Hendrickson et al., 1990). MAD data were collected from one selenomethionyl crystal held at 21708C at the HHMI beamline X4A
(NSLS, Brookhaven, NY) using Fuji imaging plates. The crystal was
aligned with its c-axis parallel with v (the oscillation axis) to facilitate
the simultaneous collection of anomalous pairs. The four wavelengths correspond to the inflection, the peak, and two remote values of the X-ray absorption spectrum of the derivatized crystal (l2,
l3, l1 , and l4, respectively, Table 1). Data were processed using
DENZO and SCALEPACK (Z. Otwinowski and W. Minor). Six of the
ten possible selenium sites were found using SHELXS-90 in Patterson search mode (Sheldrick, 1991). The other 4 sites were found
from difference Fouriers. Phasing was treated as a MIRAS problem,
all 10 selenium sites were refined and phases calculated to 2.8 Å
using MLPHARE (Z. Otwinowski) with three of the wavelengths
treated as derivatives and one (l2) treated as parent. The resulting
map revealed clear delineation between protein and solvent and a
few helices were discernable (Figure 6). A dramatically improved
map was generated using SQUASH (Zhang and Main, 1990), into
which a polyalanine model of z50% of the residues was built using
the program O (Jones et al., 1991). The map was improved by cycles
of refinement using X-PLOR (Brünger, 1992), phase combination
using SIGMAA (Read, 1986), and model building. None of the nonglycine backbone torsion angle combinations lie in unfavorable regions
of the Ramachandran plot; 92% lie in the most favored regions.
Coordinates will be submitted to the Brookhaven Protein Data Bank.
Cell
134
Figure 6. Electron Density Maps
Stereo views of (upper panel) the experimental, nonsolvent flattened electron density map (2.8 Å, 1.0 s), and (lower panel) the 2|Fo| 2 |Fc |
electron density map (2.6 Å, 1.3 s) calculated from the final refined coordinates, shown as atomic stick figures using color coding for atom
type (C, yellow; O, red; N, blue; two water molecules are shown as red crosses), generated using the program O (Jones et al., 1991). Solventexposed aromatic residues surrounded by basic residues from conserved region 2.3, proposed to be involved in promoter melting (Juang
and Helmann, 1994; deHaseth and Helmann, 1995), are displayed.
Structure of a s 70 Subunit Fragment
135
Acknowledgments
(1991). Three-dimensional structure of yeast RNA polymerase II at
16 Å resolution. Cell 66, 121–128.
We thank J. Kuriyan and S. K. Burley and the members of their
laboratories for help and encouragement, and in particular J. Goldberg for invaluable advice. We thank W. A. Hendrickson and C.
Ogata for access to and support at beamline X4A, which is funded
by the Howard Hughes Medical Institute at the Brookhaven National
Laboratory. We also thank W. Minor for help with DENZO and M.
Lonetto for communicating unpublished sequence alignment information. S. A. D. is a Lucille P. Markey Scholar and a Pew Scholar
in the Biomedical Sciences. This work was supported in part by
grants to S. A. D. from the Lucille P. Markey Charitable Trust, the
Irma T. Hirschl Trust, the Human Frontier Science Project, the Pew
Foundation, and the National Institutes of Health.
deHaseth, P.L., and Helmann, J.D. (1995). Open complex formation
by Escherichia coli RNA polymerase: The mechanism of polymerase-induced strand separation of double helical DNA. Mol. Microbiol. 16, 817–824.
Received July 19, 1996; revised August 13, 1996.
References
Ahearn, J.M., Bartolomei, M.S., West, M.L., Cisek, L.J., and Corden,
J.L. (1987). Cloning and sequence analysis of the mouse genomic
locus encoding the largest subunit of RNA polymerase II. J. Biol.
Chem. 262, 10695–10705.
Aiyar, S.E., Juang, Y.L., Helmann, J.D., and deHaseth, P.L. (1994).
Mutations in sigma factor that affect the temperature dependence
of transcription from a promoter, but not from a mismatch bubble
in double-stranded DNA. Biochemistry 33, 11501–11506.
Allison, L.A., Moyle, M., Shales, M., and Ingles, C.J. (1985). Extensive
homology among the largest subunits of eukaryotic and prokaryotic
RNA polymerases. Cell 42, 599–610.
Amouyal, M., and Buc, H. (1987). Topological unwinding of strong
and weak promoters by RNA polymerase. A comparison between
the lac wild-type and the UV5 sites of Escherichia coli. J. Mol. Biol.
195, 795–808.
Biggs, J., Searles, L.L., and Greenleaf, A.L. (1985). Structure of the
eukaryotic transcription apparatus: Features of the gene for the
largest subunit of Drosophila RNA polymerase II. Cell 42, 611–621.
Brünger, A. (1992). X-PLOR (Version 3.1) Manual. 260 Whitney Avenue, New Haven, CT 06511, The Howard Hughes Medical Institute
and Department of Molecular Biophysics and Biochemistry, Yale
University.
Buckle, M., and Buc, H. (1989). Fine mapping of DNA single-stranded
regions using base-specific chemical probes. Biochemistry 28,
4388–4396.
Buckle, M., Geiselmann, J., Kolb, A., and Buc, H. (1991). Protein-DNA
cross-linking at the lac promoter. Nucleic Acids Res. 19, 833–840.
Burgess, R.R., Travers, R.R., Dunn, J.J., and Bautz, E.K. F. (1969).
Factor stimulating transcription by RNA polymerase. Nature 221,
43–44.
Dombroski, A.J., Walter, W.A., Record, M.T., Siegele, D.A., and
Gross, C.A. (1992). Polypeptides containing highly conserved regions of transcription initiation factor sigma70 exhibit specificity of
binding to promoter DNA. Cell 70, 501–512.
Dombroski, A.J., Walter, W.A., and Gross, C.A. (1993). Amino-terminal amino acids modulate sigma-factor DNA-binding activity. Genes
Dev. 7, 2446–2455.
Gardella, T., Moyle, T., and Susskind, M.M. (1989). A mutant Escherichia coli sigma 70 subunit of RNA polymerase with altered promoter specificity. J. Mol. Biol. 206, 579–590.
Gribskov, M., and Burgess, R.R. (1986). Sigma factors from E. coli,
B. subtilis, phase SPO1, and phage T4 are homologous proteins.
Nucleic Acids Res. 14, 6745–6763.
Gross, C.A., Lonetto, M., and Losick, R. (1992). Bacterial sigma
factors. Transcriptional regulation Eds. K. Yamamoto and S.
McKnight. Cold Spring Harbor, New York, Cold Spring Harbor Laboratory.
Ha, I., Lane, W.S., and Reinberg, D. (1991). Cloning of a human gene
encoding the general transcription initiation factor IIB. Nature 352,
689–694.
Harley, C.B., and Reynolds, R.P. (1987). Analysis of E. coli promoter
sequences. Nucleic Acids Res. 15, 2343–2361.
Hawley, D.K., and McClure, W.R. (1983). Compilation and analysis
of Escherichia coli promoter DNA sequences. Nucleic Acids Res.
11, 2237–2255.
Helmann, J.D., and Chamberlin, M.J. (1988). Structure and function
of bacterial sigma factors. Ann. Rev. Biochem. 57, 839–872.
Hendrickson, W., Norton, J.R., and LeMaster, D.M. (1990). Selenomethionyl proteins produced for analysis by multiwavelength anomalous diffraction (MAD). EMBO J. 9, 1665–1672.
Hendrickson, W.A. (1991). Determination of macromolecular structures from anomalous diffraction of synchrotron radiation. Science
254, 51–58.
Hilton, M.D., and Whiteley, H.R. (1985). UV cross-linking of the Bacillus subtilis RNA polymerase to DNA in promoter and non-promoter
complexes. J. Biol. Chem. 260, 8121–8117.
Hinkle, D.C., and Chamberlin, M.J. (1972). Studies of the binding of
Escherichia coli RNA polymerase to DNA. I. The role of sigma subunit
in site selection. J. Mol. Biol. 70, 157–185.
Carson, M. (1991). Ribbons 2.0. J. Appl. Crystallogr. 24, 958–961.
Horikoshi, M., Wang, C.K., Fujii, H., Cromlish, J.A., Weil, P.A., and
Roeder, R.B. (1989). Cloning and structure of a yeast gene encoding
a general transcription initiation factor TFIID that binds to the TATA
box. Nature 341, 299–303.
Chan, B., Minchin, S., and Busby, S. (1990). Unwinding of duplex
DNA during transcription initiation at the Escherichia coli galactose
operon overlapping promoters. FEBS Lett. 267, 46–50.
Jang, S.H., and Jaehning, J.A. (1991). The yeast mitochondrial RNA
polymerase specificity factor, MTF1, is similar to bacterial s factors.
J. Biol. Chem. 266, 22671–22677.
Christie, G.E., and Cale, S.B. (1995). The sigma-70 subunit from
Escherichia coli C differs from that of E. coli K-12. Gene 162,
161–162.
Jones, C.H., and Moran, C.P.J. (1992). Mutant s factor blocks transition between promoter binding and initiation of transcription. Proc.
Natl. Acad. Sci. USA 89, 1958–1962.
Conaway, J.W., and Conaway, R.C. (1990). An RNA polymerase II
transcription factor shares functional properties with Escherichia
coli sigma 70. Science 248, 1550–1553.
Jones, T.A., Zou, J.-Y., Cowan, S., and Kjeldgaard, M. (1991). Improved methods for building protein models in electron denstity
maps and the location of errors in these models. Acta Crystallogr.
A47, 110–119.
Conaway, R.C., and Conaway, J.W. (1993). General initiation factors
for RNA polymerase II. Ann. Rev. Biochem. 62, 161–190.
Daniels, D., Zuber, P., and Losick, R. (1990). Two amino acids in an
RNA polymerase s factor involved in the recognition of adjacent
base pairs in the 210 region of a cognate promoter. Proc. Natl.
Acad. Sci. USA 87, 8075–8079.
Juang, Y.-L., and Helmann, J.D. (1994). A promoter melting region
in the primary sigma factor of Bacillus subtilis: identification of functionally important aromatic amino acids. J. Mol. Biol. 235, 1470–
1488.
Darst, S.A., Kubalek, E.W., and Kornberg, R.D. (1989). Three-dimensional structure of Escherichia coli RNA polymerase holoenzyme
determined by electron crystallography. Nature 340, 730–732.
Juang, Y.-L., and Helmann, J.D. (1995). Pathway of promoter melting
by Bacillus subtilis RNA polymerase at a stable RNA promoter:
effects of temperature, d protein, and s factor mutations. Biochemistry 34, 8465–8473.
Darst, S.A., Edwards, A.M., Kubalek, E.W., and Kornberg, R.D.
Kenney, T.J., York, K., Youngman, P., and Moran, C.P.J. (1989).
Cell
136
Genetic evidence that RNA polymerase associated with s A factor
uses a sporulation-specific promoter in Bacillus subtilis. Proc. Natl.
Acad. Sci. USA 86, 9109–9113.
Kirkegaard, K., Buc, H., Spassky, A., and Wang, J.C. (1983). Mapping
of single-stranded regions in duplex DNA at the sequence level:
single-strand-specific cytosine methylation in RNA polymerase-promoter complexes. Proc. Natl. Acad. Sci. USA 80, 2544–2548.
Lesley, S.A., Brow, M.A.D., and Burgess, R.R. (1991). Use of in vitro
protein synthesis from polymerase chain reaction-generated templates to study interactions of Escherichia coli transcription factors
with core RNA polymerase and for epitope mapping of monoclonal
antibodies. J. Biol. Chem. 266, 2632–2638.
Lesley, S.A., and Burgess, R.R. (1989). Characterization of the Escherichia coli transcription factor sigma 70: localization of a region
involved in the interaction with core RNA polymerase. Biochemistry
28, 7728–7734.
Lonetto, M., Gribskov, M., and Gross, C.A. (1992). The s 70 family:
sequence conservation and evolutionary relationships. J. Bacteriol.
174, 3843–3849.
polymerase interacts homologously with two different promoters.
Cell 20, 269–281.
Siegele, D.A., Hu, J.C., Walter, W.A., and Gross, C.A. (1989). Altered
promoter recognition by mutant forms of the sigma 70 subunit of
Escherichia coli RNA polymerase. J. Mol. Biol. 206, 591–603.
Simpson, R.B. (1979). The molecular topology of RNA polymerasepromoter interaction. Cell 18, 277–285.
Sopta, M., Burton, Z.F., and Greenblatt, J. (1989). Structure and
associated DNA-helicase activity of a general transcription initiation
factor that binds to RNA polymerase II. Nature 341, 410–414.
Spassky, A., Rimsky, S., Buc, H., and Busby, S. (1988). Correlation
between the conformation of Escherichia coli 210 hexamer sequences and promoter strength: use of orthophenanthroline cuprous complex as a structural index. EMBO J. 7, 1871–1879.
Stragier, P., Parsot, C., and Bouvier, J. (1985). Two functional domains conserved in major and alternate bacterial sigma factors.
FEBS Lett. 187, 11–15.
Losick, R., and Pero, J. (1981). Cascades of sigma factors. Cell 25,
582–584.
Sumimoto, H., Ohkuma, Y., Sinn, E., Kato, H., Shimasaki, S., Horikoshi, M., and Roeder, R.G. (1991). Conserved sequence motifs in
the small subunit of human general transcription factor TFIIE. Nature
354, 401–404.
McCracken, S., and Greenblatt, J. (1991). Related RNA polymerasebinding regions in human RAP30/74 and Escherichia coli s 70. Science 253, 900–902.
Sweetser, D., Nonet, M., and Young, R.A. (1987). Prokaryotic and
eukaryotic RNA polymerases have homologous core subunits. Proc.
Natl. Acad. Sci. USA 84, 1192–1196.
Nagai, K., Oubridge, C., Jessen, T.H., Li, J., and Evans, P.R. (1990).
Crystal structure of the RNA-binding domain of the U1 small nuclear
ribonucleoprotein A. Nature 348, 515–520.
Tatti, K.M., Jones, C.H., and Moran, C.P.J. (1991). Genetic evidence
for interaction of sigma E with the spoIIID promoter in Bacillus
subtilis. J. Bacteriol. 173, 7828–7833.
Nicholls, A., Sharp, K.A., and Honig, B. (1991). Protein folding and
association: insights from the interfacial and thermodynamic properties of hydrocarbons. Proteins Struct. Funct. Genet. 11, 281–296.
Tintut, Y., and Gralla, J.D. (1995). PCR mutagenesis identifies a
polymerase-binding sequence of sigma 54 that includes a sigma 70
homology region. J. Bacteriol. 177, 5818–5825.
Nikolov, D.B., Hu, S.H., Lin, J., Gasch, A., Hoffmann, A., Horikoshi,
M., Chua, N.H., Roeder, R.G., and Burley, S.K. (1992). Crystal structure of TFIID TATA-box binding protein. Nature 360, 40–46.
Travers, A.A. (1990). Why bend DNA? Cell 60, 177–180.
Ohkuma, Y., Sumimoto, H., Hoffmann, A., Shimasaki, S., Horikoshi,
M., and Roeder, R.G. (1991). Structural motifs and potential s homologies in the large subunit of human general transcription factor
TFIIE. Nature 354, 398–401.
Waldburger, C., Gardella, T., Wong, R., and Susskind, M.M. (1990).
Changes in conserved region 2 of Escherichia coli sigma 70 affecting
promoter recognition. J. Mol. Biol. 215, 267–276.
Park, C.S., Hillel, Z., and Wu, C.-W. (1980). DNA strand specificity
in promoter recognition by RNA polymerase. Nucleic Acids Res. 8,
5895–5912.
Polyakov, A., Severinova, E., and Darst, S.A. (1995). Three-dimensional structure of Escherichia coli core RNA polymerase: promoter
binding and elongation conformations of the enzyme. Cell 83,
365–373.
Read, R. (1986). Improved Fourier coefficients for maps using
phases from partial structures with errors. Acta Crystallogr. A42,
140–149.
Roeder, R.G. (1991). The complexities of eukaryotic transcription
initiation: regulation of preinitiation complex assembly. Trends Biochem. Sci. 16, 402–408.
Rong, J.C., and Helmann, J.D. (1994). Genetic and physiological
studies of Bacillus subtilis sA mutants defective in promoter melting.
J. Bacteriol. 176, 5218–5224.
Schultz, P., Celia, H., Riva, M., Sentenac, A., and Oudet, P. (1993).
Three-dimensional model of yeast RNA polymerase I determined
by electron microscopy of two-dimensional crystals. EMBO J. 12,
2601–2607.
Shamoo, Y., Friedman, A.M., Parsons, M.R., Konigsberg, W.H., and
Steitz, T.A. (1995). Crystal structure of a replication fork singlestranded DNA binding protein (T4 gp32) complexed to DNA. Nature
376, 362–366.
Sheldrick, G.M. (1991). Isomorphous Replacement and Anomalous
Scattering, In Proc. CCP4 St. W., W. Wolf, P.R. Evans, and A.G.W.
Leslie, eds. (Warrington, UK: Daresbury Laboratory), 23–28.
Shuler, M.F., Tatti, K.M., Wade, K.H., and Moran, C.P.J. (1995). A
single amino acid substitution in s E affects its ability to bind core
RNA polymerase. J. Bacteriol. 177, 3687–3694.
Siebenlist, U., Simpson, R.B., and Gilbert, W. (1980). E. coli RNA
Travers, A.A., and Burgess, R.R. (1969). Cyclic re-use of the RNA
polymerase sigma factor. Nature 222, 537–540.
Wong, C., Tintut, Y., and Gralla, J.D. (1994). The domain structure
of sigma 54 as determined by analysis of a set of deletion mutants.
J. Mol. Biol. 236, 81–90.
Zhang, K.Y.J., and Main, P. (1990). Histogram matching as a new
density modification technique for phase refinement and extension
of protein molecules. Acta Crystallogr. A46, 41–46.
Zuber, P., Healy, J., Carter, H.L., III, Cutting, S., Moran, C.P., Jr.,
and Losick, R. (1989). Mutation changing the specificity of an RNA
polymerase sigma factor. J. Mol. Biol. 206, 605–614.
Note Added in Proof
The data referred to above as E.S. et al., submitted, are now in
press as the following: Severinova, E., Severinov, K., Fenyö, D.,
Marr, M., Brody, E.N., Roberts, J.W., Chait, B.T., and Darst, S.A.
(1996). Domain organization of the Escherichia coli RNA polymerase
s70 subunit. J. Microbiol., in press.